No abstract

The stability and persistence of coral reefs in the decades to come is uncertain due to global warming and repeated bleaching events that will lead to reduced resilience of these ecological and socio-economically important ecosystems. Identifying key refugia is potentially important for future conservation actions. We suggest that the Gulf of Aqaba (GoA) (Red Sea) may serve as a reef refugium due to a unique suite of environmental conditions. Our hypothesis is based on experimental detection of an exceptionally high bleaching threshold of northern Red Sea corals and on the potential dispersal of coral planulae larvae through a selective thermal barrier estimated using an ocean model. We propose that millennia of natural selection in the form of a thermal barrier at the southernmost end of the Red Sea have selected coral genotypes that are less susceptible to thermal stress in the northern Red Sea, delaying bleaching events in the GoA by at least a century.

Abstract. A box model of the ocean-atmosphere-sea ice-land ice climate system is used to study a novel mechanism for self-sustained oscillations of the climate system on a time scale of 100,000 years, without external forcing. The oscillation in land ice volume has the familiar sawtooth shape of climate proxy records. The most novel aspect of the climate oscillations analyzed here is the crucial role played by the sea ice. The sea ice acts as a "switch" of the climate system, switching it from a growing land glaciers mode to a retreating land glaciers mode and shaping the oscillation's sawtooth structure. A simple explanation of the 100-kyr timescale is formulated on the basis of the mechanism seen in the model. Finally, rapid sea ice changes such as those seen in our model, and their drastic effects on the climate system, may provide an explanation to some of the rapid climate changes observed to be a part of the variability at all timescales in the paleorecord.

[1] A mechanism is proposed for the mid-Pleistocene transition from a dominant periodicity of 41 kyr to 100 kyr in glacial oscillations. The same mechanism is shown to also explain the asymmetry between the long glaciation and short deglaciation phases of each cycle since that transition versus the symmetry of the 41-kyr oscillations prior to the transition. These features arise naturally within the framework of the sea-ice switch glacial cycle mechanism of Gildor and Tziperman [2000] as a result of the gradual cooling of the deep ocean during the Pleistocene. This cooling results in a change of the relation between atmospheric temperature and the rates of accumulation and ablation of continentals ice sheets. It is this latter change that leads to the activation of the sea-ice switch and therefore to the initiation of the 100-kyr oscillations. The gradual glaciation and rapid deglaciations during these oscillations occur because the mean value of the ice sheet ablation is not far from the maximum rate of snow accumulation during warm periods. This proximity of mean ablation and maximum accumulation rates is shown to be also a consequence of the mid-Pleistocene gradual cooling of the deep ocean.